Many people feel DRaaS is the next wave in data protection. Just like flash solves so many problems in the primary storage world, DRaaS solves a ton of problems in the backup and DR worlds. And it does all that while reducing costs. So what could be wrong with it?
It’s not so much there’s something wrong with DRaaS. The question is better stated, “Is DRaaS as good as we can do?” If you’re using the cloud as a way to store backup data, what about other copies of data? Can you use the cloud as part of a copy data management (CDM) system where backup and DR are simply one part of the puzzle?
For those unfamiliar with the concept of CDM, here’s a primer. It’s based on the idea that copies of primary data are used for more than just backup and DR. They also are used for development, test, and big data purposes. CDM advocates believe it is possible to have one copy of each version of each file or database, and simply use that copy for multiple purposes. In this context, the question is are we stepping forward in one area and backward in another when we look at moving to DRaaS? Are we creating better and faster backups and disaster recoveries, but creating yet another copy of data?
Putting this in the CDM context, try this question: is it possible to use the copy of data that we are currently using for DRaaS purposes to accomplish other goals? Can we use the copies created by DRaaS products to do testing and development? Can we do data mining against these copies for big data purposes? Even if it were made possible, will the performance of data stored in the cloud be sufficient to meet the needs of the test and development communities? Unfortunately, the answer to these questions is no for most products available today. While it is theoretically possible, it is not yet possible to do it with most products available today.
Leveraging DRaaS for CDM is another way we can reduce costs, and that’s what we talk about in our on demand webinar, “How to Cut Disaster Recovery Expenses – Improve Recovery Times”.